The long-term objective of this project is the discovery of an anti-cancer drug lead derived from the microbial polyketide, leptomycin B (LMB). This natural product has unusually potent cytotoxic activity resulting from inhibition of the nuclear export of proteins and RNA through inactivation of the exportin, CRMI. Many cell signaling processes, including ones essential to the specific onset and development of cancer, rely on CRM1 mediated nuclear export. Consequently, inhibition of CRM1 by LMB has potential applicability to the treatment a wide variety of cancers involving the normal or aberrant function of proteins such as p53, COX-2, and Bcr-Abl. Identification of a lead compound that maintains the potency of LMB but displays an improved therapeutic index will set the stage for preclinical development. To achieve this goal, we will seek an improved understanding of the structure activity relationships (SAR) for LMB and inhibition of CRM1 function and nuclear export. Our working hypothesis imagines that LMB can be divided into two portions: (i) a recognition domain for initial protein binding and (ii) a warhead domain that reversibly covalently modifies CRMI. This concept will guide experimental investigations of the SAR through pursuit of the following Specific Aims: 1) Establishment of a CRM1 filter binding assay as well as a CRMl-functional assay; 2) Semisynthesis of 18 different classes of LMB recognition domain and warhead domain analogs from LMB itself or from a truncated form obtained by polyketide synthase gene engineering and fermentation; 3) Evaluation of the biological properties of all analogs for CRM1 binding, inhibition of a CRMl-dependent functional assay, in vitro cytotoxicity in a panel of four cancer cell lines, and in vivo efficacy in specific mouse xenograph models of human cancers.